Depressive symptoms in the elderly are associated with syncope, falls, and mortality. These associations are independent of comorbid conditions, and may reflect central nervous system effects. We have been investigating insular pathology as a cause of similar problems in Alzheimer's disease (AD). AD pathology reaches the insula at a preclinical stage, which may affect 40% of septuagenarians. Insular AD lesions, particularly on the right, are associated with bradyarrhythmias, Qtc intervals, and mortality. Moreover, right hemisphere metabolism and visuospatial skills predict mortality in stroke, and epilepsy as well as AD. However, we can now show that visuospatial cognitive impairments also mediate the effect of depressive symptoms on mortality. This implicates right hemisphere dysfunction in depression's mortality risk. Thus, the diproportionate risk of depression related mortality in the elderly may be mediated by their increased risk of subclinical structural insular lesions, including stroke, Lewy Body and /or subclinical AD pathology. This study will determine whether assymetric reductions in right insular blood flow by 3T magnetic resonance imaging (MRI) mediates the associaiton between depressive symptoms and measures of autonomic function, which have previously been associated with mortality in elderly populations. Our Primary Outcome will be the ratio of low frequency: high frequency components in the spectral distribution of Holter derived R-R frequency variations. Additonal heart rate variability (HRV) measures and longitudinal changes in heart rate HRV will be used in Secondary Analyses. Exploratory analyses will address other MRI pathological markers as potential mediators. Significant mediation would have important public health implications. 38% of community dwelling elderly with Mild Cognitive Impairment and could be at risk on the basis of subclinical insular AD pathology. 48% of middle cerebral artery strokes affect the insula. Minorities are disporportionately at risk for combined AD pathology and depressive symptoms. Moreover, medications may interact with insular pathology to affect mortality risk. Psychotropics, for example, affect insular blood flow. This may explain recent reports of increased mortality in eldely patients receiving atypical neuroleptics or cholinesterase inhibitors. Even anti-depressants may aggravate this risk, as they do not reliably improve depression related changes in HRV, and are themselves independent predictors of syncope, falls and hip fracture. Simple screening tests, such as clock- drawing, may identify those at greatest risk. PUBLIC RELEVANCE: Depressive symptoms in the elderly are associated with fainting spells, falls, bone fractures, and death. This study will determine whether right insular dysfunction may be a cause of these problems. Simple drawing tests may identify those at risk. These may include as many as 38% of community dwelling elderly with Mild Cognitive Impairment (MCI), or 24% of persons with middle cerebral artery strokes. Minorities may be disporportionately at risk. Moreover, insular dysfunction could explain recent reports of increased mortality in eldely patients receiving several types of psychotropic medications.